Project 1: Synaptic potentiation in D2 receptor expressing neurons in the accumbens confers protection against the development of compulsive cocaine use. (Bock, Shin et al.; under revision) The hypothesis behind this study is that the vulnerability to develop compulsive cocaine use arises in part from inherent differences across animals in the ability to recruit the D2-MSNs and engage the indirect accumbal-tegmental pathway. Evidence from previous studies suggests that activation of D2 receptors is involved in the locomotor response to cocaine and conditioned place preference (Durieux et al., 2009; Ferguson et al., 2011). However, it is still unclear what the role of D2-MSNs is with regard to voluntary cocaine self-administration and the development of compulsive cocaine use. Experiments also test the hypothesis that inhibition of D2-MSNs impairs self-control over cocaine intake and renders individuals more susceptible to the rewarding effects of the drug. The results showed that potentiation of excitatory inputs onto D2-MSNs occurs only in mice that can exert self-control over cocaine intake, and suggest that the potentiation might confer protection. Furthermore, the experiments suggested that inhibition of D2-MSN output renders mice vulnerable to the expression of compulsive behaviors by enhancing the motivation to obtain cocaine, without affecting drug use when it is readily available. These data are in agreement with previous studies showing that reinstatement of cocaine seeking is associated with reduced extracellular GABA in the ventral pallidum (Tang et al., 2005). In conclusion, this study establishes synaptic potentiation in D2-MSN inputs as a critical mechanism for controlling the expression of compulsive behaviors towards cocaine. We propose that this cell-specific synaptic potentiation facilitates the recruitment of indirect pathway neurons and protects against the development addictive behaviors. This study also showed that manipulations of D2-MSN activity, even when introduced after several weeks of drug taking, can successfully alter the motivation to obtain cocaine. The use of therapeutic brain stimulation to silence ventral pallidum neurons or to selectively activate D2-MSNs could enhance self-control in patients fighting dependence on drugs of abuse. Project 2: Acute cocaine effects on glutamatergic and dopaminergic transmission in the nucleus accumbens (Adrover, Shin and Alvarez, in preparation) Glutamatergic inputs to the NAc that originate in the VTA have been recently demonstrated and recent studies suggest that glutamatergic transmission from VTA DA neurons plays an important role in the rewarding properties of cocaine and amphetamine (Birgner et al., 2010; Kozorovitskiy et al., 2012), as well as enhanced sucrose and cocaine self-administration (Alsio et al., 2011). However, the properties of these synapses are poorly understood and the effect of cocaine unknown. In this study, optogenic tools are used to selectively activate terminals from VTA DA neurons into the NAc and investigate the acute effect of cocaine on DA and glutamatergic transmission. ChR2 activation in the NAc shell region was achieved by a brief laser pulse and triggered DA transients that were recorded with a carbon fiber using fast scanning cyclic voltammetry (FSCV). DA transients evoked by ChR2 activation also had similar properties to those evoked by electrical stimulation. This same ChR2 activation evoked glutamatergic excitatory postsynaptic currents (EPSCs) in MSNs of the shell that were abolished by TTX but were insensitive to D1R and D2R antagonists. The average amplitude of the AMPA receptor-mediated EPSC was 101 9 pA and the average AMPA/NMDA ratio was 1.5 0.2. Bath-application of cocaine (10 M) prolonged the DA transients and increasing the decay time constant by a factor of 10 (3.5 0.6 s). The area of the DA transient was increased 2.5 times and the peak was transiently enhanced. In contrast, cocaine inhibited AMPA-R mediated EPSCamplitude by half. A similar pattern of depression was observed in the NMDA-R mediated EPSCs.The D2R antagonist sulpiride reversed the inhibition of both AMPA-R and NMDA-R mediated EPSCs when applied after cocaine. These data indicated that D2R activation is required for cocaine-induced inhibition of EPSCs and indeed, a D2-like agonist inhibited the DA transients and also glutamatergic EPSCs in a concentration dependent manner and with a similar IC50, indicating that VTA terminals that release glutamate also express presynaptic D2Rs. D2R are expressed by indirect pathway MSNs, cholinergic interneurons and also on presynaptic DA terminals where they inhibit DA release. Mice lacking D2R only in DA neurons were also tested. Cocaine increased DA transients by 50% and had no effect on AMPA-R EPSCs in these mice. Furthermore, the DA transient response to cocaine was similar to those of wild-type mice when cocaine was added in the presence of sulpiride. These results indicate that acute cocaine, in addition to causing larger DA transients, acts on presynaptic D2R to inhibit DA and glutamate release from VTA terminals. Furthermore, these results suggest that cocaine, acting via these mechanisms, has opposite effects on DA and glutamatergic transmission from midbrain DA neurons, enhancing the first and depressing the latter. Project 3: Dissection of the drinking behavior and analysis of spine morphology in a mouse model of binge-like ethanol drinking (Wilcox et al., under review) Research into the neurobiology of heavy and binge-like ethanol drinking has been limited by the low-levels of voluntary ethanol consumption shown by most mouse strains (Crabbe et al., 2011). Recently, a model of intermittent access to ethanol has been shown to elicit binge-like drinking and pharmacologically relevant blood ethanol concentrations (BECs) in mice (Rhodes et al., 2005). Termed Drinking in the Dark, this model takes advantage of the circadian patterns of mice to achieve reliably high levels of consumption in a two hour drinking session. C57BL/6J mice reach BECs higher than 80 mg/dl, and show signs of intoxication such as motor impairment (Rhodes et al., 2007). DID is a robust paradigm that has been successfully used to investigate neuronal circuits and signals that modulate binge-like ethanol consumption (Sprow, 2012). Despite the success of this intermittent access model, the mechanisms underlying the acquisition of voluntary ethanol drinking are not completely understood. In this study, we established DID in our laboratory and showed that it produces reliable escalation of voluntary ethanol intake and blood ethanol concentration. We characterized the drinking pattern of mice with intermittent access to ethanol using lickometers to record each bout with high temporal resolution over many weeks of voluntary ethanol consumption, and analyzed the synaptic morphology of striatal neurons 2 days and 30 days after the last ethanol binge. The results represent a novel and important contribution to the alcohol field because they identify the bottle exchange, an integral part of DID, as a likely mechanism by which intermittent access facilitates the acquisition of voluntary ethanol drinking behavior by inducing mice to drink at a higher rate. Over time, mice change their ethanol drinking pattern and increase the speed of drinking at the beginning of each DID session. Faster ethanol drinking is associated with higher BEC, and an enhanced preference for ethanol was observed after 6 weeks of treatment. Interestingly, no changes were detected in either striatal or accumbal spine density, and a shortening of spine length was seen only transiently, suggesting that these behavioral changes occur independent of long-term changes in synaptic morphology in brain regions involved in the reward pathway.